Can Test Boosters Reboot?

Andy Smarick has half of a great post at the Fordham Institute blog.

Smarick suggests that the opt-out movement is testing the ed reform movement, and that the movement is not achieving a Proficient in humility. He catalogues many of the dismissive, condescending responses that have been written to opting out, seeing them as analogous to the dismissive responses to common-core standards pushback all those year ago.

But I'm concerned that education reform's propensity for pride may have taken an even more unfortunate turn with opt-out. One emerging narrative is that we should be suspicious when certain groups of people question our policy preferences.

That's a fair read. From Amanda Ripley to Merryl Tisch, the message being broadcast is that people who question the awesomeness of the Big Standardized Test are fools, racists, and hysteria-addled loons. Smarick suggests that this is not conducive to dialogue (though many reformsters like Tisch don't seem particularly interested in dialogue), and there might be a better response.

We could disparage them. But that would only serve to insult and incite.

Or we could humbly listen, respectfully argue our case, and make the necessary course corrections.

And up to that point, he's really onto something. But I'd like to suggest something more. I think the opt-out explosion suggests that more is needed than a tweaking. I'd like to suggest a reboot.

Policy reboots are great things. I reboot some portion of my classroom every year. Instead of just getting out last year's materials, I go back to the beginning and ask, "What do I want to accomplish? What would be the best way to do that?" Sometimes I reach the exact same conclusion that I did in previous years; sometimes I do not. Often I now know things that I didn't know when I first chose my approach, and so rebooting helps me find a better path. I use this as a teaching thing, but in general I think it's a bad life principle to do something today just because you did it yesterday.

So I'm going to pretend for a moment that I accept all of the stated goals of the high-stakes testing regimen, and I'm just going to go back to the beginning for each. If this is my goal, what would be the best way to pursue it?

Inform Instruction
 
The idea here is that data will help me decide in my classroom how best to change what I do. Data will also help my building principal and superintendent decide how to adjust and improve curriculum.
Right off the bat, I notice that I need two different data streams. My classroom data needs to be swift and granular. I need to know how all my students did today so that I can plan what to do tomorrow. But the school and district data don't need to be so granular—in fact, too much detail will make it too easy to lose the forest while staring at the hairs on an aphid's back. The school and district also need to decide whether they A) want to fine-tune the instruction for a particular cohort of students and adjust to meet their needs, essentially creating instructional change that follows those students like a wave or B) fine-tune a system that sits in place as each cohort passes through it. One requires data for customization while the other requires data for One Size Fits All. 

So a large-scale, national-level test may not be my best choice for informing instruction at all.
This is particularly true when you remember that one policy change that absolutely nobody has suggested is shortening the school year. Yet by using a testing system that eats up days and weeks of time, that is exactly what we've done. I would look for assessment methods that would not require us to reduce instructional time.

Making Failure Visible
 
One of our goals is to find failing schools and rescue them. To do that, we'd need to agree on what "failing" and "rescue" mean. Until we can agree on what success is supposed to look like in a school, we have no idea how to collect that information. 

However, here's one thing I believe is true about failing schools; none of them are a secret. In 10-plus years of test-driven accountability, have we ever found a school that turned out to be terrible, but nobody knew it until the test results came out? Some activists welcome the tests from a belief that the government will have to listen to them when they say poor schools are in need of resources and assistance. Maybe we could just work it out so the government listened to them. 

The best way to get a picture of a school is to listen to the people who know the school--parents, teachers, alumni, community members. If I wanted to know how a school was doing, I'd talk to those people--not just look at the narrow results of a standardized test. I don't believe that failing schools and the students struggling through them are invisible;I just don't think people in power were ever really looking (not until there was money to be made by "rescuing" them). 

Evaluating Teachers
 
I've mapped out a system for evaluating teachers; you can find a fuller picture of my plan here. What I definitely would not do is try to evaluate some teachers by testing students they don't have in subjects they don't teach.

Provide Parents With Information About Student Achievement
 
Why would I look for a proxy to tell me how my child is doing in school when I can look directly at how my child is doing in school. I would call for all manner of transparency, from public access to data about the district to opening avenues for parent-teacher communication, from the high tech of cyber-communications to the low tech of evening office hours for working parents. Transparency is not achieved by an extra test unrelated to the usual business of the school, particularly not when the contents of that test are a secret. 

Taxpayer Accountability
 
Again, we first need to know what the taxpayers want the school system to do. Ask 100 citizens what they consider most important in a school, and I don't believe that any number will answer, "I don't care what else you do as long as those kids score well on a Big Standardized Test." Taxpayers should expect and receive accountability, but coming up with the instrument will be difficult. Again, the cheapest and most direct means would be transparency;let people see as much as possible of what's going on behind the walls (without violating student privacy). 

It occurs to me that pro-test folks might think I'm engaging in some sort of rhetorical ju-jitsu here, but let me assure you, I am not. One thing that continues to amaze and intrigue me about the test-driven accountability movement is that these folks have a list of Very Important Things they want to accomplish, and they have somehow focused on an instrument that is a lousy way to accomplish any of them. It's like watching someone who says, "You must prepare me a gourmet meal or else!" and then hands the cook a can of Spam, a hammer, and a box of matches. 

There are discussions worth having about the value of some of these goals, but if these really are your goals, in high-stakes standardized tests you have chosen the exactly wrong instrument.
Educators have been telling you this for over a decade. Parents and students are now telling you as well. It is time to step back and consider what would truly be the best ways to achieve your stated goals. Because Big Standardized Tests are not it.

Originally posted at View from the Cheap Seats